Objective To automatically segment diabetic retinal exudation features from deep learning color fundus images. Methods An applied study. The method of this study is based on the U-shaped network model of the Indian Diabetic Retinopathy Image Dataset (IDRID) dataset, introduces deep residual convolution into the encoding and decoding stages, which can effectively extract seepage depth features, solve overfitting and feature interference problems, and improve the model's feature expression ability and lightweight performance. In addition, by introducing an improved context extraction module, the model can capture a wider range of feature information, enhance the perception ability of retinal lesions, and perform excellently in capturing small details and blurred edges. Finally, the introduction of convolutional triple attention mechanism allows the model to automatically learn feature weights, focus on important features, and extract useful information from multiple scales. Accuracy, recall, Dice coefficient, accuracy and sensitivity were used to evaluate the ability of the model to detect and segment the automatic retinal exudation features of diabetic patients in color fundus images. Results After applying this method, the accuracy, recall, dice coefficient, accuracy and sensitivity of the improved model on the IDRID dataset reached 81.56%, 99.54%, 69.32%, 65.36% and 78.33%, respectively. Compared with the original model, the accuracy and Dice index of the improved model are increased by 2.35% , 3.35% respectively. Conclusion The segmentation method based on U-shaped network can automatically detect and segment the retinal exudation features of fundus images of diabetic patients, which is of great significance for assisting doctors to diagnose diseases more accurately.
In the clinical stage, suspected hemolytic plasma may cause hemolysis illness, manifesting as symptoms such as heart failure, severe anemia, etc. Applying a deep learning method to plasma images significantly improves recognition accuracy, so that this paper proposes a plasma quality detection model based on improved “You Only Look Once” 5th version (YOLOv5). Then the model presented in this paper and the evaluation system were introduced into the plasma datasets, and the average accuracy of the final classification reached 98.7%. The results of this paper's experiment were obtained through the combination of several key algorithm modules including omni-dimensional dynamic convolution, pooling with separable kernel attention, residual bi-fusion feature pyramid network, and re-parameterization convolution. The method of this paper obtains the feature information of spatial mapping efficiently, and enhances the average recognition accuracy of plasma quality detection. This paper presents a high-efficiency detection method for plasma images, aiming to provide a practical approach to prevent hemolysis illnesses caused by external factors.
Although attention plays an important role in cognitive and perception, there is no simple way to measure one's attention abilities. We identified that the strength of brain functional network in sustained attention task can be used as the physiological indicator to predict behavioral performance. Behavioral and electroencephalogram (EEG) data from 14 subjects during three force control tasks were collected in this paper. The reciprocal of the product of force tolerance and variance were used to calculate the score of behavioral performance. EEG data were used to construct brain network connectivity by wavelet coherence method and then correlation analysis between each edge in connectivity matrices and behavioral score was performed. The linear regression model combined those with significantly correlated network connections into physiological indicator to predict participant's performance on three force control tasks, all of which had correlation coefficients greater than 0.7. These results indicate that brain functional network strength can provide a widely applicable biomarker for sustained attention tasks.
In audiovisual emotion recognition, representational learning is a research direction receiving considerable attention, and the key lies in constructing effective affective representations with both consistency and variability. However, there are still many challenges to accurately realize affective representations. For this reason, in this paper we proposed a cross-modal audiovisual recognition model based on a multi-head cross-attention mechanism. The model achieved fused feature and modality alignment through a multi-head cross-attention architecture, and adopted a segmented training strategy to cope with the modality missing problem. In addition, a unimodal auxiliary loss task was designed and shared parameters were used in order to preserve the independent information of each modality. Ultimately, the model achieved macro and micro F1 scores of 84.5% and 88.2%, respectively, on the crowdsourced annotated multimodal emotion dataset of actor performances (CREMA-D). The model in this paper can effectively capture intra- and inter-modal feature representations of audio and video modalities, and successfully solves the unity problem of the unimodal and multimodal emotion recognition frameworks, which provides a brand-new solution to the audiovisual emotion recognition.
Selective attention promotes the perception of brain to outside world and coordinates the allocation of limited brain resources. It is a cognitive process which relies on the neural activities of attention-related brain network. As one of the important forms of brain activities, neural oscillations are closely related to selective attention. In recent years, the relationship between selective attention and neural oscillations has become a hot issue. The new method that using external rhythmic stimuli to influence neural oscillations, i.e., neural entrainment, provides a promising approach to investigate the relationship between selective attention and neural oscillations. Moreover, it provides a new method to diagnose and even to treat the attention dysfunction. This paper reviewed the research status on the relationship between selective attention and neural oscillations, and focused on the application prospects of neural entrainment in revealing this relationship and diagnosing, even treating the attention dysfunction.
Glaucoma is one of blind causing diseases. The cup-to-disc ratio is the main basis for glaucoma screening. Therefore, it is of great significance to precisely segment the optic cup and disc. In this article, an optic cup and disc segmentation model based on the linear attention and dual attention is proposed. Firstly, the region of interest is located and cropped according to the characteristics of the optic disc. Secondly, linear attention residual network-34 (ResNet-34) is introduced as a feature extraction network. Finally, channel and spatial dual attention weights are generated by the linear attention output features, which are used to calibrate feature map in the decoder to obtain the optic cup and disc segmentation image. Experimental results show that the intersection over union of the optic disc and cup in Retinal Image Dataset for Optic Nerve Head Segmentation (DRISHTI-GS) dataset are 0.962 3 and 0.856 4, respectively, and the intersection over union of the optic disc and cup in retinal image database for optic nerve evaluation (RIM-ONE-V3) are 0.956 3 and 0.784 4, respectively. The proposed model is better than the comparison algorithm and has certain medical value in the early screening of glaucoma. In addition, this article uses knowledge distillation technology to generate two smaller models, which is beneficial to apply the models to embedded device.
This paper aims to assist the individual clinical diagnosis of children with attention-deficit/hyperactivity disorder using electroencephalogram signal detection method. Firstly, in our experiments, we obtained and studied the electroencephalogram signals from fourteen attention-deficit/hyperactivity disorder children and sixteen typically developing children during the classic interference control task of Simon-spatial Stroop, and we completed electroencephalogram data preprocessing including filtering, segmentation, removal of artifacts and so on. Secondly, we selected the subset electroencephalogram electrodes using principal component analysis (PCA) method, and we collected the common channels of the optimal electrodes which occurrence rates were more than 90% in each kind of stimulation. We then extracted the latency (200~450 ms) mean amplitude features of the common electrodes. Finally, we used the k-nearest neighbor (KNN) classifier based on Euclidean distance and the support vector machine (SVM) classifier based on radial basis kernel function to classify. From the experiment, at the same kind of interference control task, the attention-deficit/hyperactivity disorder children showed lower correct response rates and longer reaction time. The N2 emerged in prefrontal cortex while P2 presented in the inferior parietal area when all kinds of stimuli demonstrated. Meanwhile, the children with attention-deficit/hyperactivity disorder exhibited markedly reduced N2 and P2 amplitude compared to typically developing children. KNN resulted in better classification accuracy than SVM classifier, and the best classification rate was 89.29% in StI task. The results showed that the electroencephalogram signals were different in the brain regions of prefrontal cortex and inferior parietal cortex between attention-deficit/hyperactivity disorder and typically developing children during the interference control task, which provided a scientific basis for the clinical diagnosis of attention-deficit/hyperactivity disorder individuals.
Leukemia is a common, multiple and dangerous blood disease, whose early diagnosis and treatment are very important. At present, the diagnosis of leukemia heavily relies on morphological examination of blood cell images by pathologists, which is tedious and time-consuming. Meanwhile, the diagnostic results are highly subjective, which may lead to misdiagnosis and missed diagnosis. To address the gap above, we proposed an improved Vision Transformer model for blood cell recognition. First, a faster R-CNN network was used to locate and extract individual blood cell slices from original images. Then, we split the single-cell image into multiple image patches and put them into the encoder layer for feature extraction. Based on the self-attention mechanism of the Transformer, we proposed a sparse attention module which could focus on the discriminative parts of blood cell images and improve the fine-grained feature representation ability of the model. Finally, a contrastive loss function was adopted to further increase the inter-class difference and intra-class consistency of the extracted features. Experimental results showed that the proposed module outperformed the other approaches and significantly improved the accuracy to 91.96% on the Munich single-cell morphological dataset of leukocytes, which is expected to provide a reference for physicians’ clinical diagnosis.
Cranial defects may result from clinical brain tumor surgery or accidental trauma. The defect skulls require hand-designed skull implants to repair. The edge of the skull implant needs to be accurately matched to the boundary of the skull wound with various defects. For the manual design of cranial implants, it is time-consuming and technically demanding, and the accuracy is low. Therefore, an informer residual attention U-Net (IRA-Unet) for the automatic design of three-dimensional (3D) skull implants was proposed in this paper. Informer was applied from the field of natural language processing to the field of computer vision for attention extraction. Informer attention can extract attention and make the model focus more on the location of the skull defect. Informer attention can also reduce the computation and parameter count from N2 to log(N). Furthermore,the informer residual attention is constructed. The informer attention and the residual are combined and placed in the position of the model close to the output layer. Thus, the model can select and synthesize the global receptive field and local information to improve the model accuracy and speed up the model convergence. In this paper, the open data set of the AutoImplant 2020 was used for training and testing, and the effects of direct and indirect acquisition of skull implants on the results were compared and analyzed in the experimental part. The experimental results show that the performance of the model is robust on the test set of 110 cases fromAutoImplant 2020. The Dice coefficient and Hausdorff distance are 0.940 4 and 3.686 6, respectively. The proposed model reduces the resources required to run the model while maintaining the accuracy of the cranial implant shape, and effectively assists the surgeon in automating the design of efficient cranial repair, thereby improving the quality of the patient’s postoperative recovery.
Although deep learning plays an important role in cell nucleus segmentation, it still faces problems such as difficulty in extracting subtle features and blurring of nucleus edges in pathological diagnosis. Aiming at the above problems, a nuclear segmentation network combined with attention mechanism is proposed. The network uses UNet network as the basic structure and the depth separable residual (DSRC) module as the feature encoding to avoid losing the boundary information of the cell nucleus. The feature decoding uses the coordinate attention (CA) to enhance the long-range distance in the feature space and highlights the key information of the nuclear position. Finally, the semantics information fusion (SIF) module integrates the feature of deep and shallow layers to improve the segmentation effect. The experiments were performed on the 2018 data science bowl (DSB2018) dataset and the triple negative breast cancer (TNBC) dataset. For the two datasets, the accuracy of the proposed method was 92.01% and 89.80%, the sensitivity was 90.09% and 91.10%, and the mean intersection over union was 89.01% and 89.12%, respectively. The experimental results show that the proposed method can effectively segment the subtle regions of the nucleus, improve the segmentation accuracy, and provide a reliable basis for clinical diagnosis.